在我們生活的這個世界中，有很多人為的、天然的美景，我們為了欣賞這些景像，會走出家門觀看，並且經常喜歡使用相機、手機拍照，讓這些美景可以保存下來，甚至是分享到網路上，供其他人欣賞，所以不管是在網路、電視上都能輕易看到這些美景。但對於某些影像，由於受到環境的影響，造成影像模糊不清、低對比度的狀況。而影像品質不佳的原因有很多面向，像是光線不足、空氣中有雜質、水下的折射等，但不管是什麼原因引起的，我們都希望把這些照片變成更為清楚、好看。
而針對影像低對比度的狀況，本論文基於自動色彩均衡(Automatic Color Equalization)[1]演算法，提出一個有效提升影像品質的方法。自動色彩均衡是針對人眼對色彩與亮度的感知，對圖片進行修正，所以能有效把影像的色彩度、對比度以及影像品質提升，使人眼能夠更清楚明顯看到修正完後的影像。自動色彩均衡演算法可使用於個別處理影像中RGB三個Channel，或是先將RGB轉成HSV，再對V進行運算，最後將V與H、S合併後轉回RGB。自動色彩均衡演算法是針對Channel做運算，而因為每個Channel的運算涉及將當前像素與其他所有像素進行差分比較累加，所以複雜度頗高，當圖片愈大，運算時間也愈可觀，因此限制了它的應用。本論文是以自動色彩均衡為基礎，對該演算法進行部分的優化，使修改後演算法與原始演算法的效果雷同，但能使運算時間大幅下降，並設計出一個能夠有效提升該演算法執行速度的硬體架構。

In the world we live in, there are many man-made and natural beauties. In order to appreciate these scenes, we will go out to watch them, and often like to take pictures with cameras and mobile phones, so that these beauties can be preserved and even shared on the Internet for others to enjoy, so whether it is on the Internet or on TV these beautiful scenery can be easily seen. However, for some images, due to the influence of the environment, the image is blurred and low contrast. There are many reasons for poor image quality, such as lack of light, impurities in the air, refraction underwater, etc, but no matter what the reason is, we want to make these photos clearer and better.
In view of the low contrast of the image, this paper proposes an effective method to improve the image quality based on the Automatic Color Equalization [1] algorithm. Automatic Color Equalization is based on the human eye's perception of color and brightness, and corrects the image, so it can effectively improve the color, contrast and image quality of the image, so that the human eye can see the corrected image more clearly. The Automatic Color Equalization algorithm can be used to individually process the three channels of RGB in the image, or convert RGB to HSV first, then operate on V, and finally combine V with H, S and convert back to RGB. The Automatic Color Equalization algorithm operates on Channels, and because the operation of each Channel involves comparing and accumulating the current pixel and all other pixels, so the complexity is quite high, the larger the image, the more considerable the computing time, thus limiting its application. This paper optimizes part of the Automatic Color Equalization algorithm, so that the effect of the modified algorithm is similar to the original algorithm, but the operation time can be greatly reduced, and design a hardware architecture that can effectively improve the execution speed of the algorithm.

論文審定書 i
論文提要 ii
誌謝 iii
摘要 iv
Abstract v
目錄 vii
圖次 ix
表次 xi
第一章 序論 1
1.1 研究動機和目的 1
1.2 論文架構 3
第二章 研究背景 4
2.1色彩模型 4
2.1.1 RGB色彩模型 4
2.1.2 YUV色彩模型 5
2.1.3 HSV色彩模型 6
2.2 影像強化方法 8
2.2.1直方圖均衡化 8
2.2.2限制對比度自適應直方圖均衡化 9
2.2.3 深度學習CycleGAN 10
2.2.4 灰度世界演算法 12
2.2.5自動白平衡 12
2.2.6 Retinex影像強化理論 13
2.2.6.1單尺度SSR 14
2.2.6.2多尺度MSR 16
2.2.6.3帶色彩恢復的多尺度MSR 17
第三章 研究方法 18
3.1運算流程 18
3.2 自動色彩均衡 19
3.2.1色彩／空域重構 21
3.2.2動態擴展 24
3.2.3演算法步驟調整 24
第四章 提出的影像強化硬體架構 27
4.1 硬體架構 27
4.2 重構圖片硬體 28
4.2.1整體重構圖片硬體 28
4.2.2 PE (Processing Element) 30
4.3 動態擴展硬體 31
第五章 實驗結果和分析 32
5.1 實驗步驟與方法 32
5.2軟硬體實作效果比較 33
5.3水下影像強化比較結果 47
5.4硬體分析 53
第六章 結論與未來工作 57
6.1 結論 57
6.2 未來研究方向 58
參考文獻 59

[1]C. Gatta, A. Rizzi, D. Marini, “ACE: An automatic color equalization algorithm,” Proceedings of the First European Conference on Color in Graphics Image and Vision , 2002.
[2] N. Ul Sadad, A. Afrin and M. N. I. Mondal, “FPGA based Histogram Equalization for Image Processing,” 2021 3rd International Conference on Electrical & Electronic Engineering , pp. 89-92,2021. doi: 10.1109/ICEEE54059.2021.9718837
[3] Zuiderveld and Karel, “Contrast Limited Adaptive Histogram Equalization,”
Graphics Gems IV, Academic Press Professional, Inc., pp.474-485, 1994.
[4] A. S. Parihar and K. Singh, “A study on Retinex based method for image enhancement, ” 2018 2nd International Conference on Inventive Systems and Control , pp. 619-624,2018. doi: 10.1109/ICISC.2018.8398874
[5] W. Steve, “Digital Compositing for Film and Video,” Focal Press,ISBN
024080760X,2006.
[6] M. Joe, “RGB and YUV Color,” FXScript Reference, 2003.
[7] A. R. Smith, “Color Gamut Transform Pairs,” ACM Siggraph Computer Graphics,
vol.12, no.3, pp.12-19, 1978.
[8] I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, Y. Bengio, “ Generative Adversarial Networks, ” arXiv preprint arXiv:1406.2661, 2014.
[9] J. Zhu, T. Park, P. Isola and A. A. Efros, “Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks, ” 2017 IEEE International Conference on Computer Vision ,2017. pp. 2242-2251, doi: 10.1109/ICCV.2017.244
[10]C. Li,J. Guo, C. Guo, “ Emerging From Water: Underwater Image Color Correction Based on Weakly Supervised Color Transfer, ” arXiv preprint arXiv: 1710.07084, 2017.
[11] E. Provenzi, C. Gatta, M. Fierro and A. Rizzi, “ A Spatially Variant White-Patch and Gray-World Method for Color Image Enhancement Driven by Local Contrast, ” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 30, no. 10, pp. 1757-1770,2008. doi: 10.1109/TPAMI.2007.70827
[12] L. EH, ‘‘An alternative technique for the computation of the designator in the retinex theory of color vision,’’ Proc Natl Acad Sci USA,83(10):3078-80,1986. doi: 10.1073/pnas.83.10.3078. PMID: 3458165; PMCID: PMC323455
[13] https://blog.csdn.net/ajianyingxiaoqinghan/article/details/71435098
[14] Z. Rahman, D. J. Jobson and G. A. Woodell, ‘‘Multi-scale retinex for color image enhancement,’’ Proceedings of 3rd IEEE International Conference on Image Processing, pp. 1003-1006 vol.3,1996. doi: 10.1109/ICIP.1996.560995
[15] D. J. Jobson, Z. Rahman and G. A. Woodell, ‘‘A multiscale retinex for bridging the gap between color images and the human observation of scenes,’’ in IEEE Transactions on Image Processing, vol. 6, no. 7, pp. 965-976,1997. doi: 10.1109/83.597272
[16] https://www.cxymm.net/article/lilingyu520/46755767
[17] M. Yang and A. Sowmya, “An Underwater Color Image Quality Evaluation Metric,” IEEE Transactions on Image Processing, vol. 24, no. 12, pp.6062–6071, 2015.
[18] K. Panetta, C. Gao, and S. Agaian, “Human-Visual-System-Inspired Underwater Image Quality Measures,” IEEE Journal of Oceanic Engineering, vol. 41, no. 3, pp. 541–551, 2016.
[19] A. Mittal, R. Soundararajan, and A. C. Bovik, “Making a "Completely Blind" Image Quality Analyzer,” IEEE Signal Processing Letters, vol. 20, no. 3, pp. 209–212, 2013
[20] J. A. Palacios, V. Caro, M. Durán and M. Figueroa, ‘‘A hardware architecture for Multiscale Retinex with Chromacity Preservation on an FPGA,’’ Euromicro Conference on Digital System Design, pp. 73-80,2020. doi: 10.1109/DSD51259.2020.00023
[21] Y.-T. Peng and P. C. Cosman, “Underwater image restoration based on image blurriness and light absorption,” IEEE transactions on image processing, vol. 26, no. 4, pp. 1579–1594, 2017.
[22] P. L. Drews, E. R. Nascimento, S. S. Botelho, and M. F. M. Campos, “Underwater depth estimation and image restoration based on single images,” IEEE computer graphics and applications, vol. 36, no. 2, pp. 24–35, 2016.
[23] H. Li, J. Li, and W. Wang, “A fusion adversarial underwater image enhancement network with a public test dataset,” arXiv preprint arXiv:1906.06819, 2019.
[24] C. Li, J. Guo, and C. Guo, “Emerging from water: Underwater image color correction based on weakly supervised color transfer,” IEEE Signal Processing Letters, vol. 25, no. 3, pp. 323–327, 2018.
[25] J.-Y. Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image-to-image translation using cycle-consistent adversarial networks,” IEEE International Conference on Cmputer Vision, pp. 2223–2232,2017.
[26] J. S. Romero, L. M. Procel, L. Trojman and D. Verdier, ‘‘Implementation and optimization of the algorithm of automatic color enhancement in digital images,’’ IEEE International Autumn Meeting on Power, Electronics and Computing,2017. pp. 1-6. doi: 10.1109/ROPEC.2017.8261651
[27] A. R. S. M. and S. M. H., ‘‘Underwater image enhancement using single scale retinex on a reconfigurable hardware,’’ International Symposium on Ocean Electronics , pp. 1-5,2015. doi: 10.1109/SYMPOL.2015.7581166
[28] Y. Li, H. Zhang, Y. You, and M. Sun, ‘‘A multi-scale Retinex implementation on FPGA for an outdoor application,’’ in Proc. Congr. Image Signal Process, pp. 1788–1792,2011.
[29] https://blog.csdn.net/weixin_36474809/article/details/88778213